Make Better Decisions, Leverage New Opportunities, and Automate Decisioning at ScalePrescriptive analytics is more directly linked to successful decision-making than any other form of business analytics. It can help you systematically sort through your choices to optimize decisions, respond to new opportunities and risks with precision, and continually reflect new information into your decisioning process.In Prescriptive Analytics, analytics expert Dr. Dursun Delen illuminates the field’s state-of-the-art methods, offering holistic insight for both professionals and students. Delen’s end-to-end, all-inclusive approach covers optimization, simulation, multi-criteria decision-making methods, inference- and heuristic-based decisioning, and more. Balancing theory and practice, he presents intuitive conceptual illustrations, realistic example problems, and real-world case studies–all designed to deliver knowledge you can use.
Chapter 1 Introduction to Business Analytics and Decision-Making 1Data and Business Analytics 1An Overview of the Human Decision-Making Process 4 Simon’s Theory of Decision-Making 5An Overview of Business Analytics 21 Why the Sudden Popularity of Analytics? 22 What Are the Application Areas of Analytics? 23 What Are the Main Challenges of Analytics? 24A Longitudinal View of Analytics 27A Simple Taxonomy for Analytics 31Analytics Success Story: UPS’s ORION Project 36 Background 37 Development of ORION 38 Results 39 Summary 40Analytics Success Story: Man Versus Machine 40 Checkers 41 Chess 41 Jeopardy! 42 Go 42 IBM Watson Explained 43Conclusion 47References 47Chapter 2 Optimization and Optimal Decision-Making 49Common Problem Types for LP Solution 51Types of Optimization Models 52 Linear Programming 52 Integer and Mixed-Integer Programming 52 Nonlinear Programming 53 Stochastic Programming 54Linear Programming for Optimization 55 LP Assumptions 56 Components of an LP Model 58 Process of Developing an LP Model 59 Hands-On Example: Product Mix Problem 60 Formulating and Solving the Same Product-Mix Problem in Microsoft Excel 68 Sensitivity Analysis in LP 72Transportation Problem 76 Hands-On Example: Transportation Cost Minimization Problem 76 Network Models 81Hands-On Example: The Shortest Path Problem 82 Optimization Modeling Terminology 89Heuristic Optimization with Genetic Algorithms 92 Terminology of Genetic Algorithms 93 How Do Genetic Algorithms Work? 95 Limitations of Genetic Algorithms 97 Genetic Algorithm Applications 98Conclusion 98References 99Chapter 3 Simulation Modeling for Decision-Making 101Simulation Is Based on a Model of the System 106What Is a Good Simulation Application? 110Applications of Simulation Modeling 111Simulation Development Process 113 Conceptual Design 114 Input Analysis 114 Model Development, Verification, and Validation 115 Output Analysis and Experimentation 116Different Types of Simulation 116 Simulation May Be Dynamic (Time-Dependent) or Static (Time-Independent) 117 Simulations May Be Stochastic or Deterministic 118 Simulations May Be Discrete and Continuous 118Monte Carlo Simulation 119 Simulating Two-Dice Rolls 120 Process of Developing a Monte Carlo Simulation 122 Illustrative Example–A Business Planning Scenario 125 Advantages of Using Monte Carlo Simulation 129 Disadvantages of Monte Carlo Simulation 129Discrete Event Simulation 130 DES Modeling of a Simple System 131 How Does DES Work? 135 DES Terminology 138System Dynamics 143Other Varieties of Simulation Models 149 Lookahead Simulation 149 Visual Interactive Simulation Modeling 150 Agent-Based Simulation 151Advantages of Simulation Modeling 153Disadvantages of Simulation Modeling 154Simulation Software 155Conclusion 158References 159Chapter 4 Multi-Criteria Decision-Making 161Types of Decisions 164A Taxonomy of MCDM Methods 165 Weighted Sum Model 170 Hands-On Example: Which Location Is the Best for Our Next Retail Store? 172Analytic Hierarchy Process 173 How to Perform AHP: The Process of AHP 176 AHP for Group Decision-Making 184 Hands-On Example: Buying a New Car/SUV 185Analytics Network Process 190 How to Conduct ANP: The Process of Performing ANP 194Other MCDM Methods 201 TOPSIS 202 ELECTRE 202 PROMETHEE 204 MACBETH 205Fuzzy Logic for Imprecise Reasoning 207 Illustrative Example: Fuzzy Set for a Tall Person 208Conclusion 210References 210Chapter 5 Decisioning Systems 213Artificial Intelligence and Expert Systems for Decision-Making 214An Overview of Expert Systems 222 Experts 222 Expertise 223 Common Characteristics of ES 224Applications of Expert Systems 228 Classical Applications of ES 228 Newer Applications of ES 229Structure of an Expert System 232 Knowledge Base 233 Inference Engine 233 User Interface 234 Blackboard (Workplace) 234 Explanation Subsystem (Justifier) 235 Knowledge-Refining System 235Knowledge Engineering Process 236 1 Knowledge Acquisition 237 2 Knowledge Verification and Validation 239 3 Knowledge Representation 240 4 Inferencing 241 5 Explanation and Justification 247Benefits and Limitations of ES 249 Benefits of Using ES 249 Limitations and Shortcomings of ES 253 Critical Success Factors for ES 254Case-Based Reasoning 255 The Basic Idea of CBR 255 The Concept of a Case in CBR 257 The Process of CBR 258 Example: Loan Evaluation Using CBR 260 Benefits and Usability of CBR 260 Issues and Applications of CBR 261Conclusion 266References 267Chapter 6 The Future of Business Analytics 269Big Data Analytics 270 Where Does the Big Data Come From? 271 The Vs That Define Big Data 273 Fundamental Concepts of Big Data 276 Big Data Technologies 280 Data Scientist 282 Big Data and Stream Analytics 284Deep Learning 289 An Introduction to Deep Learning 291 Deep Neural Networks 295 Convolutional Neural Networks 296 Recurrent Networks and Long Short-Term Memory Networks 301 Computer Frameworks for Implementation of Deep Learning 304Cognitive Computing 308 How Does Cognitive Computing Work? 310 How Does Cognitive Computing Differ from AI? 311Conclusion 312References 313Index 315
Dr. Dursun Delen is an internationally known expert in business analytics and data mining. He is often invited to national and international conferences to deliver keynote presentations on topics related to data/text mining, business intelligence, decision support systems, business analytics, and knowledge management. Prior to his appointment as professor at Oklahoma State University in 2001, Dr. Delen worked for industry for more than 10 years, developing and delivering business analytics solutions to companies. Most recently he worked for a privately owned research and consulting company, Knowledge Based Systems, Inc., in College Station, Texas, as a research scientist. During his five years there, he led a number of projects related to decision support, information systems, and advanced analytics that were funded by federal agencies, including the DoD, NASA, NIST, and the DOE. Today, in addition to his academic endeavors, Dr. Delen provides consulting services to businesses in assessing their information systems needs and developing state-of-the-art business analytics capabilities.